Identification and validation of impact models

Research output: Contribution to conferenceAbstractAcademic

Abstract

Predicting the outcome of an impact event is of high importance for proper execution of many robotic tasks. Rigid-body
contact models are extensively used in planning and control due to their simplicity and computational efficiency. However, there exists
little literature that shows a comparison and verification of these models with real-life experiments, and it is therefore unclear how
well these models approximate frictional impact events. In this study, we formulate an identification approach to find the parameters
of commonly used contact, friction, and impact models using an experimentally obtained data-set of impact events using a single
rigid-body, where the focus lies specifically on spatial frictional impact events. In future work, we will measure, using the identified
parameters, the performance of the used models in terms of long-horizon prediction performance by comparing simulated and measured
rest poses of a rigid-body tossed on a surface.
Original languageEnglish
Number of pages2
Publication statusPublished - 18 Jul 2022
Event10th European Nonlinear Dynamics Conference, ENOC 2022 - Université de Lyon, Lyon, France
Duration: 17 Jul 202222 Jul 2022
Conference number: 10
https://enoc2020.sciencesconf.org/

Conference

Conference10th European Nonlinear Dynamics Conference, ENOC 2022
Abbreviated titleENOC 2022
Country/TerritoryFrance
CityLyon
Period17/07/2222/07/22
OtherENOC 2020:
Due to COVID-19 pandemia, the conference is postponed to 17-22 July 2022 at the same location.
Internet address

Fingerprint

Dive into the research topics of 'Identification and validation of impact models'. Together they form a unique fingerprint.

Cite this